. The long term health effects of exposure to environmental contaminants are major public concerns. Occupational or environmental exposure to radiation, fuel hydrocarbons, and polychlorinated biphenyls is associated with increased risk of malignancy, particularly leukemia. However, assessment of the genotoxic and carcinogenic potential of environmental contaminants is difficult. Uncertainties in physical exposure and variations in physiologic response among individuals complicates risk assessment. Cross-species data extrapolation from in vitro and in vivo animal model systems is problematic. Fortunately, recent advances in molecular and cellular methodologies and in vivo models promise to alleviate some of these limitations. Molecular cytogenetics facilitates detection of translocations using fluorescence in situ hybridization. Immune -deficient mice transplanted with human cord blood cells (HU-mice) maintain high levels of persistent multi-lineage hematopoiesis, thereby providing opportunities to evaluate molecular and functional changes in human hemopoietic cells exposed to genotoxins in vivo. HU-mice should allow assessment of genotoxin effects on human hemopoietic stem cells (hsc) in vivo. We propose a novel approach to assess genomic and cellular changes in human hsc exposed to environmental contaminants in vivo. specifically, we will determine whether human hsc in HU-mice exposed to single and intermittent low doses of radiation or chemicals provide a read-out assay useful to estimate exposure. Specifically, we will 1) determine whether the frequency of genetically-damaged human hsc in an optimized HU-mouse model reflects in vitro exposure to radiation or benzene. 2) establish the relationship between in vivo radiation and benzene exposure and frequency of genetically damaged human hsc. 3) Test the hypothesis that intermittent, low dose in vivo exposure of human has to a genotoxin with concomitant hematotoxicity will result in more genomic damage than dose-equivalent single exposures. These studies will establish the usefulness of the HU-mouse model to increase understanding of the relationship between genotoxin dose and genetic damage in human hsc- an important component of risk assessment.